Audit Trail

How Fund Analyst Intelligence records what changed, when, why, and who approved it, with reproducible monthly cycles and evidence-linked outputs.

Audit Trail

Fund Analyst Intelligence is designed to be defensible after the fact.
A month later, a quarter later, or during an audit, the same question appears.
What changed, when did it change, why did we accept it, and who approved it?

The audit trail answers that question with a production record.
It is not a narrative explanation.
It is a structured, reproducible history.

Objectives

A production audit trail must ensure:

  • every cycle has a complete record of inputs and outputs
  • every change is linked to evidence
  • every reviewer decision is attributable
  • every published report is reproducible
  • exceptions and follow-ups are traceable to closure

What is recorded

Fund Analyst Intelligence records audit data at four layers.

1. Cycle record

Each monthly cycle is a first-class entity.

Cycle record includes

  • cycle identifier and timestamps
  • fund identifier and scope version
  • run status and completion markers
  • operator and reviewer roles
  • applied materiality policy version
  • applied validation rule-set version

This makes every run explainable.
It also makes changes in policy visible.

2. Input inventory

Every cycle stores a controlled inventory of sources.

Includes

  • artefact ids and versions
  • ingestion timestamps
  • source origin and metadata
  • online retrieval timestamps where applicable
  • coverage notes (what is missing or stale)

This prevents reconstructing “what we used” from memory.

3. Change and decision log

The audit trail must capture both deltas and human decisions.

Includes

  • structured deltas vs prior snapshot
  • change category labels
  • evidence links per change
  • exception lifecycle events
  • reviewer actions: accept, edit, reject, follow-up, monitor
  • rationale notes for overrides and resolutions

This is the core accountability layer.

4. Output registry

Every published artefact is registered.

Includes

  • monthly memo and quarterly pack identifiers
  • generation timestamp and template version
  • approval stamp and publication state
  • associated evidence pack reference
  • reproducibility pointer to the cycle record

Outputs are treated as production artefacts.
They are not “files produced somewhere”.

Snapshot history

The platform maintains snapshot history per fund.
A snapshot represents the validated state at a point in time.

Snapshots enable

  • deterministic delta computation
  • quarter-over-quarter comparisons
  • trend analysis and recurring issue detection
  • reproducible reporting from historical states

A snapshot is only created when the cycle is approved.
This prevents polluted histories from draft states.

What “reproducible” means

A report is reproducible if you can:

  • identify the cycle that produced it
  • retrieve the exact input artefacts and versions
  • re-run the validation and delta logic deterministically
  • apply the stored reviewer decisions and overrides
  • regenerate the output template using the stored version

Fund Analyst Intelligence is designed around cycle records and snapshots to make this possible.

Review and approval accountability

Accountability is enforced through explicit states.

Typical approval requirements

  • cycle marked complete only after reviewer sign-off
  • publication blocked unless approval state is “approved”
  • overrides require rationale notes for key fields
  • follow-ups must have an owner and status

This protects teams from silent drift.
It also protects the organisation under scrutiny.

Follow-ups and closure

Audit is not only about reports.
It is also about unresolved items.

Fund Analyst Intelligence tracks:

  • open exceptions and their age
  • follow-up owners and next actions
  • escalation events
  • closure decisions and supporting evidence

This ensures “known issues” are not lost across cycles.

What auditors and committees can ask, confidently

With a production audit trail, you can answer:

  • Which sources were used this month for fund X?
  • What changed since last cycle, and what evidence supports it?
  • Which exceptions were identified, and how were they resolved?
  • Who approved the final report, and when?
  • Why was a manual override applied, and what was the rationale?
  • Can you reproduce the report that was sent three months ago?

The platform is designed to make these answers routine.

Definition of done

Audit trail readiness is production-grade when:

  • each cycle has an immutable record of inputs, checks, deltas, and outputs
  • each material change links to evidence and a decision record
  • reviewer actions are attributable and time-stamped
  • snapshots update only on approval
  • reports can be reproduced from stored cycle records
  • follow-ups are tracked to closure with ownership and escalation history

This is how Fund Analyst Intelligence stays defensible at scale.